Cryptocurrency consensus algorithms

The main problem of peer-to-peer networks, especially in finance, is the so-called “Byzantine Generals” problem. To solve it, Satoshi Nakamoto suggested that users reach consensus based on complex mathematical calculations. This requires quite a lot of resources that consume an incredible amount of electricity. This is why the first cryptocurrency is now being criticized. However, developers of free funds are constantly on the way to solving this problem – how to achieve a legitimate consensus without resorting to specialized equipment that uses a lot of electricity.

Currently, there are several dozen algorithms for achieving consensus in distributed registry networks, but the most popular and efficient ones clearly deserve attention. Each algorithm has its pros and cons, but the most popular is still PoW

Proof of Work
The proof-of-work consensus algorithm is used in many popular cryptocurrencies. This is the approach Satoshi Nakamoto decided to use, and, as time showed, he was not mistaken, despite all the shortcomings, PoW remains a reliable way to keep the system working. The popular Ethereum, Litecoin, Monero and a number of others also run on it.

As mentioned earlier, in Proof of Work, miners solve a mathematical problem. This requires a large amount of energy resources to ensure the work of computing power. For a number of coins, such as Bitcoin and Ethereum, special ASIC miners have been developed. And if the second most capitalized cryptocurrency can still be mined on video cards, although not all, but having more than six gigabytes of RAM, and further this figure will only increase, the bitcoin makes sense to mine only on ASIC.

One of the features that allow to reliably solve the problem of Byzantine generals, floating complexity of the problem to be solved. This is how the network can be kept operational, eliminating the possibility of mining the entire block volume quickly.

After including all transactions in a block and sending it to the distributed registry network, the block is checked by other members of the network. This process is quite fast and does not require a lot of resources.

From the pluses of the Proof of Work algorithm we can say that with the growing popularity of the cryptocurrency, when the number of miners reaches a certain, rather high mass, the system becomes resistant to any attacks, and their conduct is inexpedient, primarily for economic reasons. But the disadvantages, unfortunately, this approach has a few more, especially in distributed registry networks, where ASIC devices are widespread:

• extremely high power consumption of the crypto-mining network, which is necessary for the sustainable operation of the entire system;
• the shortage of modern mining devices, due to which, for example, almost 70% of the total hash rate is controlled by the five largest pools, thus, there may be a risk of collusion between the latter to gain full control over the network;
• The high demand for the latest generation video cards led to a sharp more than twofold increase in prices on the retail market, which led to a shortage of older devices, little suitable for mining.

To eliminate precisely these shortcomings, developers are creating different approaches to algorithmic consensus building.

Proof of Stake
Proof of Stake is the second most popular consensus algorithm in which all the work of confirming transactions and including them in a block takes place, conventionally, in virtual space, without using a large amount of computing power. While this will solve the environmental problem of greenhouse gas emissions from power generation for mining, it is not all smooth sailing either.

Although Proof of Stake is divided into several varieties, it is based on the same principle of mining new coins. Unlike PoW, in PoS, it is not miners who confirm transactions and create new blocks, but validators. The latter block a certain amount of cryptocurrency coins as a so-called bet, and then start searching for valid blocks that can be added to the distributed registry. When such a block is found, the validator sends it to the registry, putting its coins on it. Then, after the found block is added, the network member is rewarded in the proportion in which the bet was made.

Thus, the proof-of-ownership algorithm is more convenient than the proof-of-work done. It does not require a lot of resources, a little CPU time is enough. However, PoS has one significant disadvantage: the need to have some amount of mined cryptocurrency coins in a wallet that cannot be spent. For example, in the new Ethereum protocol, code-named 2.0, it will be necessary to freeze almost $90,000 in the network at the current exchange rate. Although there are essentially no more mining costs, it makes the distributed registry network more susceptible to a 51% attack – an attacker can buy a large number of coins and then distribute them to different wallets, making them validators, then changing the registry or accepting only the blocks he wants. Thus, cryptocurrencies running on the Proof of Stake consensus algorithm are subject to greater centralization. Algorithms with some rules and restrictions have been developed to solve some proof of ownership problems.

Nothing at Stake.
This is not a subspecies of proof of ownership, but rather a solution to one of the problems – maintaining any distributed ledger chains, in general, the bet made remains with the validator regardless of whether its block was accepted into the shared blockchain or not, and whether it is of legal origin or created for malicious intent.

Ethereum 2.0 solved this problem. Just like in the general case, the validator puts some of his frozen coins as a share, and when the block is found and added, having passed the validation he will get them back together with some of the money owed to him. Otherwise, when a block is found to be malicious, the validator will be fined, and if he attempts to add the block to the roster without a stake, he will be locked out of the network entirely.

Proof of Stake Time.
In this case, it is not the number of coins in the account that counts, but the time they have been in the wallet. There will be more trust in older addresses, funds from which have not been transferred for a long period of time. In some cases, both methods can be combined – the minimum required amount of coins and the minimum time in the account.

Delegated Proof of Stake.
Delegated Proof of Stake is not exactly related to the PoS algorithm, but has the same essence. In this case, the validator does not directly look for and verify blocks, but delegates its authority to a select member of the network. The delegate is given the right to write blocks to the distributed registry. The rate in this case is a vote for the validator, which can be from 21 to 100, depending on the load of the network. If an elected delegate writes an illegitimate block to the registry, bettors vote against it and the proxy is replaced. This approach allows the network to scale to several million transactions per second. Although this approach somewhat violates the decentralized nature of cryptocurrencies, it is what makes it several times faster than other consensus algorithms in cryptocurrencies.